Comment by Gradient on Comment Submissions - Availability ... · malformations, autism spectrum disorders) for tabulation of the nickel exposure levels and study results, with no

Comments on the Hazard Identification Materials for Consideration in Listing Nickel and Nickel Compounds as Reproductive Toxicants Under Proposition 65 Submitted by Julie E. Goodman, Ph.D., DABT, ACE, ATS and Robyn L. Prueitt, Ph.D., DABT of Gradient Prepared for the Nickel Producers Environmental Research Association (NiPERA) September 10, 2018

i G:\Projects\218143_NiPERA_Prop65\WorkingFiles\Gradient_Comments_OEHHA_Nickel_Prop65_09102018.docx

Table of Contents

Page Executive Summary ES-1

1 Introduction ........................................................................................................................ 1

2 Systematic Review .............................................................................................................. 2 2.1 CalOEHHA Evidence Review .................................................................................... 2 2.2 DARTIC Review ........................................................................................................ 2 2.3 Systematic Review Approaches .............................................................................. 3

3 Epidemiology Study Quality and Results ............................................................................ 5 3.1 Nickel Forms ............................................................................................................ 5 3.2 Risk-of-Bias Analysis ................................................................................................ 6 3.3 Studies of Female Reproductive Outcomes............................................................ 7 3.4 Studies of Male Reproductive Outcomes ............................................................... 7 3.5 Studies of Developmental Outcomes ..................................................................... 8 3.6 Summary ............................................................................................................... 10

4 Conclusions ....................................................................................................................... 11

References .................................................................................................................................... 12

Tables

ii G:\Projects\218143_NiPERA_Prop65\WorkingFiles\Gradient_Comments_OEHHA_Nickel_Prop65_09102018.docx

List of Tables

Table 1 Characteristics of Epidemiology Studies Evaluating Nickel Exposure and Reproductive and Developmental Effects that Impact Study Quality

Table 2 Results of Epidemiology Studies Evaluating Nickel Exposure and Reproductive and Developmental Effects

Table 3 Risk-of-bias Criteria for Epidemiology Studies Evaluating Nickel Exposure and Reproductive and Developmental Outcomes

Table 4 Risk-of-bias Analysis of Epidemiology Studies Evaluating Nickel Exposure and Reproductive and Developmental Outcomes

ES-1 G:\Projects\218143_NiPERA_Prop65\WorkingFiles\Gradient_Comments_OEHHA_Nickel_Prop65_09102018.docx

Executive Summary

The California Office of Environmental Health Hazard Assessment (CalOEHHA) selected nickel and

nickel compounds for consideration on the Proposition 65 list of chemicals that cause reproductive toxicity.

As part of this process, CalOEHHA (2018) provided a review of the available human and experimental

animal studies evaluating associations between nickel exposure and reproductive and developmental health

outcomes to the Developmental and Reproductive Toxicant Identification Committee (DARTIC), and the

committee is expected to render an opinion regarding whether nickel and nickel compounds have been

clearly shown to cause reproductive toxicity at a meeting on October 11, 2018.

Based on the CalOEHHA (2018) evidence review, as well as a separate risk-of-bias analysis and evaluation

of the results of the same studies reviewed by CalOEHHA, we conclude the following:

While CalOEHHA conducted a substantial literature review, it did not use a systematic approach to assess the evidence. Study inclusion and exclusion criteria were not explicitly stated, study

quality was not assessed in a consistent manner, and the evidence integration sections focused only

on positive study results, without any consideration of study quality or relevance. This resulted in

an evaluation that did not fully represent the state of the science regarding the potential reproductive

and developmental toxicity of nickel and nickel compounds.

Due to the lack of a systematic approach and evaluation of study quality, reliance on the CalOEHHA evidence review will limit the ability of DARTIC to form scientifically defensible

opinions regarding the reproductive hazard potential of nickel, making it difficult for DARTIC to

determine whether nickel and nickel compounds meet the CalOEHHA criteria for listing as a

known reproductive toxicant.

To address this issue, we conducted a risk-of-bias analysis of the epidemiology studies reviewed by CalOEHHA, using the National Toxicology Program (NTP) Office of Health Assessment and

Translation (OHAT) Risk of Bias Rating Tool. The "risk of bias" of a study is the extent to which

the results are credible for any reported link between exposure and outcome, based on the design

and conduct of the study. In addition, we evaluated the results of the studies, with consideration of

how the factors that can affect the risk of bias and study quality may have impacted the

interpretation of the results. We also integrated evidence across studies, placing more weight on

higher quality studies with lower risks of bias.

Our risk-of-bias analysis found that all studies have an overall moderate risk of bias, indicating generally low quality, due to the lack of appropriate statistical approaches to assess potential

confounding, the use of area-level exposure measurements, and an inability to assess the temporal

relationship between nickel exposure and the outcome of interest.

The epidemiology studies also do not allow for an evaluation of any specific form of nickel. Oxidic and water-soluble nickel are the predominant forms of nickel found in the studies of ambient air

exposure and welders, while refinery workers are exposed to mixtures of nickel metal and soluble

and insoluble nickel compounds.

An evaluation of the results of the epidemiology studies, with consideration of how the factors that affect the risk of bias impact the interpretation of the results, indicates that the studies do not provide

evidence that nickel and/or nickel compounds are male or female reproductive or developmental

toxicants. Results across the various reproductive and developmental outcomes examined were

ES-2 G:\Projects\218143_NiPERA_Prop65\WorkingFiles\Gradient_Comments_OEHHA_Nickel_Prop65_09102018.docx

largely inconsistent or null, with no clear evidence for associations between nickel and any

particular outcome.

o Of the three studies that investigated female reproductive effects, two reported null associations. The third study reported associations for a subset of the parameters tested;

these results could be attributable to bias or confounders and have not been confirmed in

other studies. The overall evidence from human studies does not support a causal

association between nickel exposure and female reproductive effects.

o Of the eight studies that evaluated potential associations between nickel exposure and male reproductive outcomes, none accounted for important confounding variables, employed

appropriate statistical approaches, or were able to assess temporal relationships because of

their cross-sectional design. More importantly, because all the studies were found to have

a moderate risk of bias, the validity of their results is questionable. Overall, the results for

each of the male reproductive outcomes examined were inconsistent across studies, and do

not support a hazard listing for nickel and/or nickel compounds as male reproductive

toxicants.

o Twenty-eight studies evaluating potential associations between nickel exposure and various developmental outcomes, including birth defects, low birth weight, adverse

pregnancy outcomes, autism spectrum disorder (ASD), early-life cancers, and DNA

damage, were reviewed by CalOEHHA.

Seven studies evaluated nickel associations with birth defects. Two studies reported statistically significant associations with birth defects, whereas four

reported no associations and one reported a statistically significant negative (i.e.,

protective) association with nickel exposure. One of the positive studies was a

nickel refinery study for which subsequent, more thorough investigations of the

same cohort did not reproduce the positive findings. Because the majority of the

studies reported null or negative results (including those with more reliable results

as indicated by the risk-of-bias analysis), they do not support a causal association

between nickel exposure and birth defects.

Ten studies evaluated nickel associations with measures of low birth weight. Four of these studies reported statistically significant, positive associations between

nickel exposure and lower birth weight, one study reported a borderline

statistically significant association, four studies reported no associations, and one

study reported a negative (i.e., protective) association. The study reporting a

negative association had higher exposures and adequate statistical power to detect

the effects on low birth weight reported in one of the positive studies, undermining

the results of the latter study. Together, the studies of nickel and low birth weight

do not provide evidence to support a causal association.

Additional studies examining nickel exposure and ASD, early-life cancers, pneumonias, spontaneous abortion and premature birth, and DNA oxidative

damage do not provide evidence for a causal association, as each outcome was

evaluated in only a single or few studies and none of the studies accounted for

confounders.

Based on our evaluation of the risk of bias and study results, we conclude that the epidemiology studies are of generally low quality, and do not provide evidence that nickel and nickel compounds

present a reproductive or developmental hazard.

1 G:\Projects\218143_NiPERA_Prop65\WorkingFiles\Gradient_Comments_OEHHA_Nickel_Prop65_09102018.docx

1 Introduction

The California Office of Environmental Health Hazard Assessment (CalOEHHA) selected nickel and

nickel compounds for consideration on the Proposition 65 list of chemicals that cause reproductive toxicity

(which includes both reproductive and developmental toxicity). As part of this process, CalOEHHA

provided hazard identification materials to the Developmental and Reproductive Toxicant Identification

Committee (DARTIC), and the committee is expected to render an opinion regarding whether nickel and

nickel compounds have been clearly shown to cause reproductive toxicity at a meeting on October 11, 2018.

Included in these materials is a document CalOEHHA (2018) developed entitled Evidence on the

Developmental and Reproductive Toxicity of Nickel and Nickel Compounds. This document reviews the

available human and experimental animal studies evaluating associations between nickel exposure and

developmental and reproductive health outcomes.

The CalOEHHA document summarizes each study in narrative form, but does not follow a systematic

approach for evaluating study quality that is applied consistently across studies and that is considered during

the integration of the evidence. We conducted a risk-of-bias analysis of the epidemiology studies reviewed

in the CalOEHHA document, based on study quality characteristics that may have impacted the validity of

the findings. In addition, we evaluated the results of the studies, with consideration of how the factors that

can affect the risk of bias and study quality may have impacted the interpretation of the results. We also

integrated evidence across studies, placing more weight on higher quality studies with lower risks of bias,

and we considered the form of nickel to which the studied populations were likely exposed.

Based on our evaluation of the risk of bias and study results, we conclude that the epidemiology studies are

of generally low quality, and do not provide evidence that nickel or nickel compounds present a

reproductive or developmental hazard. The bases for these conclusions are described in the following

sections.


2 Systematic Review

2.1 CalOEHHA Evidence Review

In its review of the evidence for hazard identification for nickel and nickel compounds, CalOEHHA (2018)

summarized the available epidemiology and experimental animal studies evaluating associations between

nickel exposure and developmental and reproductive health outcomes.

The literature search strategy to identify relevant studies was provided in an appendix and includes a list of

the databases and search terms that were used, but there is no discussion of the decision criteria used for

study inclusion or exclusion. The document also does not discuss the number of studies identified in the

literature searches nor the steps taken to narrow down the initial list of studies to those selected for review.

The study summaries provided by CalOEHHA (2018) are in narrative form, with tables of basic study

characteristics provided in an appendix. The study narratives do not follow any consistent format, with

certain study limitations that were noted by the study authors or identified by CalOEHHA provided in a

comments section at the end of some, but not all, of the narratives. Similarly, the study tables in the

appendix provide some information on study limitations in a "Comments" column, but neither the narratives

nor the tables evaluate study quality in a manner that is consistent across all studies.

For each of the three health outcomes assessed (developmental toxicity, female reproductive toxicity, and

male reproductive toxicity), CalOEHHA (2018) conducted what it referred to as an "integrative" evaluation

of the human and animal evidence, both separately and together. For the integrative evaluations of each

realm of evidence alone, studies were grouped by similar specific outcomes (e.g. fetal growth, congenital

malformations, autism spectrum disorders) for tabulation of the nickel exposure levels and study results,

with no discussion of study quality. The integration of human and experimental animal evidence together

focuses only on the positive results of studies for specific outcomes, and does not consider study quality or

null results. There is also no comparison of the reported effect levels between experimental animals and

humans or the exposure levels between the general population and occupationally-exposed workers.

CalOEHHA (2018) did not use a systematic approach to summarize the evidence for hazard identification,

resulting in an evaluation that, even if comprehensive, lacks transparency and reproducibility and does not

fully represent the state of the science regarding the potential reproductive and developmental toxicity of

nickel compounds. As discussed below, the current evidence review does not provide a sound basis for

rendering opinions regarding the potential reproductive and developmental hazards of nickel and nickel

compounds.

2.2 DARTIC Review

The DARTIC is considering the review of the evidence by CalOEHHA (2018) to render an opinion

regarding whether nickel and nickel compounds have been clearly shown to cause reproductive toxicity.

To guide its decision, DARTIC is relying on a set of criteria for recommending chemicals for listing as

"known to the State to cause reproductive toxicity" (CalOEHHA, 1993). The criteria document states that

to be recommended for listing, there must be either sufficient evidence for developmental and reproductive


effects in humans, limited or suggestive evidence in humans supported by sufficient experimental animal

data, or sufficient evidence in experimental animals such that extrapolation to humans is appropriate.

"Sufficient" evidence in humans is defined as convincing evidence to support a causal relationship from

any of a variety of epidemiological studies "so long as the study or studies are valid according to generally

accepted principles" and the studies include "accurate exposure and toxicity endpoint classification and

proper control of confounding factors, bias, and effect modifiers" (CalOEHHA, 1993). The "weight-of-

evidence" considerations for sufficient evidence state that effects should occur in more than one human

study if the listing will be based on epidemiologic evidence alone, but that data from a single, well-

conducted study showing a clear relationship between exposure and effect may be sufficient for listing if

there are no equally well-conducted studies that do not show an effect and that "have sufficient power to

call into question the repeatability of the observation in the positive study" (CalOEHHA, 1993).

The criteria document does not define "limited" or "suggestive" evidence in humans, so it is unclear how

DARTIC is supposed to consider these criteria. Even so, the manner in which CalOEHHA (2018)

summarized the available literature for nickel compounds will likely make it difficult for DARTIC to apply

the CalOEHHA criteria for listing (CalOEHHA, 1993). This is because the study narratives and tables of

results do not allow for an evaluation of whether the epidemiology studies are scientifically valid, with

accurate exposure and outcome classification and proper control of confounding factors and bias; nor do

they allow for an evaluation of whether there is convincing evidence to support a causal relationship

between exposure to nickel and nickel compounds and developmental or reproductive effects. Ideally, the

review of the evidence should have been conducted in a consistent and reproducible way, incorporating

study quality considerations into the evidence integration process, to assist DARTIC in forming

scientifically defensible opinions.

2.3 Systematic Review Approaches

Many scientific and regulatory agencies are incorporating systematic review approaches in their evaluations

of chemical hazard and risk to minimize subjectivity and increase the transparency, rigor, and consistency

of their reviews. These include the United States Environmental Protection Agency (US EPA, 2018a,b),

the European Food Safety Authority (EFSA, 2017), the Texas Commission on Environmental Quality

(TCEQ, 2017), and the National Toxicology Program (NTP, 2015a). In contrast, CalOEHHA did not

incorporate the principles of systematic review (particularly the best practices for study selection, study

quality evaluation, and evidence integration) in its evidence review.

As noted above, CalOEHHA (2018) did not provide clear methods or decision criteria for inclusion or

exclusion of human and experimental animal studies in its evidence review for nickel compounds. This is

not consistent with the principles of transparency fundamental to systematic reviews. By not including

clear study inclusion and exclusion criteria and justification for study exclusion, it is unclear whether all

relevant studies were included for review by CalOEHHA and what the decisions for excluding studies were

based upon. Thus, it is unclear whether all pertinent information regarding the potential reproductive and

developmental toxicity of nickel compounds is available in the evidence review for consideration by

DARTIC.

CalOEHHA (2018) also did not assess study quality in a consistent manner across studies in its evidence

review of nickel compounds. Because CalOEHHA did not use a systematic approach to evaluate study

quality prior to evaluating study results, with the application of the same set of predefined study quality

criteria across all studies of the same realm (epidemiology or experimental animal), each study could not

be evaluated in an objective manner so that all study results could be considered and given appropriate

weight based on study quality rather than whether the findings were positive or null.


Another issue is that the evidence integration sections in the CalOEHHA (2018) evidence review of nickel

compounds focused on positive study results, without any consideration of study quality or relevance. In

adherence with systematic review principles, evidence reviews should include a discussion regarding how

the factors that affect study quality impact the interpretation of the results, how results from low quality

studies will be considered (particularly if they are inconsistent with results from higher quality studies), and

how null and negative study findings will be integrated into the evaluation to inform the interpretation of

positive findings. This decreases potential bias in how each of the findings is used to draw conclusions

regarding the strength of the evidence. CalOEHHA (2018) did not provide such a discussion, and it does

not appear that study quality was sufficiently considered by CalOEHHA when integrating the results of

epidemiology studies.

Evidence from human and experimental animal studies should be integrated in a manner that allows each

study type to inform the interpretation of the other. This should consider questions of human relevance,

including information on human-relevant exposures, dose-dependent effects, and species-specific

differences in toxicokinetics or susceptibility. This allows for sound judgment to be used when evaluating

whether study findings should constitute evidence for or against the hazard question. CalOEHHA (2018)

focused only on the positive results of studies in humans and experimental animals separately, and did not

compare the reported effect levels between species in its integration of the evidence.


3 Epidemiology Study Quality and Results

CalOEHHA (2018) identified 40 epidemiology studies in its evidence review for nickel and nickel

compounds; three studies evaluated female reproductive outcomes, nine studies evaluated male

reproductive outcomes, and 28 studies evaluated developmental outcomes. Study characteristics and results

for each of these studies are summarized here in Tables 1 and 2, respectively. It is notable that one male

reproductive study (Slivkova et al. 2009) and one developmental study (Huang et al., 2011) identified by

CalOEHHA did not actually evaluate statistical associations between nickel exposure and any health

outcome and should not be included in a hazard assessment of the potential reproductive toxicity of nickel

compounds.

Below, we consider the different forms of nickel to which the populations in the identified epidemiology

studies may be exposed. Then, because no specific criteria were used by CalOEHHA (2018) to evaluate

the quality of the identified studies, we conducted a standardized risk-of-bias analysis based on

epidemiology study quality characteristics that may have impacted the validity of the study findings. We

also evaluated the results of the studies, with consideration of the form of nickel evaluated and how the

factors that affect the risk of bias impact the interpretation of the results.

3.1 Nickel Forms

Humans can be exposed to many different forms of nickel in the environment. Nickel compounds can be

grouped into the four general categories of soluble, sulfidic, oxidic, and metallic nickel, with the latter three

categories consisting of compounds that are insoluble or slightly soluble (Goodman et al., 2009). In the

evidence review document for nickel, it was acknowledged that the various nickel compounds differ in their

toxicity; however, CalOEHHA (2018) incorrectly stated that the least toxic forms to humans are the soluble

nickel salts and the most toxic forms are the sulfidic and oxidic forms. While this is true for respiratory

carcinogenicity after inhalation exposure, for general toxicity it has been shown in both acute and chronic

experimental animal studies that the opposite is true; the soluble nickel salts are the most toxic, and the

insoluble nickel oxides are the least toxic (ATSDR, 2005; Goodman et al., 2011).

The majority of the epidemiology studies reviewed by CalOEHHA (2018) do not specify the nickel

compound(s) being evaluated for associations with reproductive or developmental outcomes. This is

because measurements of individual exposures to nickel compounds using biological samples (such as

blood, urine, or hair) do not differentiate among nickel forms. With regard to all the studies that evaluated

exposures to nickel in ambient air, nickel is predominantly found in suspended particulate matter (and thus

also in soil, dust, and food) in the form of both oxides and sulfates (US EPA, 1986; ATSDR, 2005;

CalOEHHA, 2012). For non-occupationally exposed study participants, the majority of nickel measured

in blood or urine is derived from the diet (De Brouwere et al., 2012). Only a few of the epidemiology

studies evaluated occupational exposures, including in nickel refineries, which can be to multiple forms of

nickel (Goodman et al., 2009), and in welders exposed to nickel and other metals. Nickel in welding fumes

is mostly present as nanometer-sized, complex metal oxides (i.e., spinels).


3.2 Risk-of-Bias Analysis

To evaluate the risk of bias for each study, we used the NTP Office of Health Assessment and Translation

(OHAT) Risk of Bias Rating Tool (NTP, 2015b), which aids in the assessment of a study's internal validity

(i.e., whether the design and conduct of the study compromised the credibility of any reported link between

exposure and outcome). The OHAT Risk of Bias Rating Tool was developed using recent guidance from

multiple organizations such as the Agency for Healthcare Research and Quality (AHRQ), Cochrane, the

CLARITY Group at McMaster University, and the Navigation Guide, and comments from the public,

technical advisors, and staff at other federal agencies (NTP, 2015b).

Using this tool, we assessed potential sources of bias using a set of questions, with detailed criteria provided

under each question that are specific for each study design (e.g., cohort, case-control, cross-sectional).

These criteria define the aspects of study design, conduct, and reporting that are used to assign a risk-of-

bias rating for each question. For epidemiology studies, the questions were divided into three key domains

(exposure assessment, outcome assessment, and confounding), as well as three other risk-of-bias domains

(selection bias, attrition bias, and statistical methods), and three domains specific to the study types and

outcomes being reviewed (exposure levels, form of nickel, and temporality). The questions and criteria are

specific enough that if different investigators applied them to the studies reviewed here, it is highly likely

that they would assign the same risk of bias ratings as we did to each study. The specific questions and

criteria for each of the nine domains are presented in Table 3.

We assigned risk-of-bias ratings to the 40 studies for each of the nine domains (Table 4). We then used the

guidance from the NTP Handbook for Conducting a Literature-based Health Assessment Using OHAT

Approach for Systematic Review and Evidence Integration (herein, the "OHAT Handbook;" NTP, 2015a)

for dividing studies into three tiers of study quality based on their risk-of-bias ratings across domains. In

this approach, studies are divided into tiers of increasing risk of bias as follows:

Tier 1 – A study must be rated as "definitely low" or "probably low" risk of bias for all key domains AND have most other risk of bias domains as "definitely low" or "probably low."

Tier 2 – Study does not meet the criteria for Tier 1 or Tier 3.

Tier 3 – A study must be rated as "definitely high" or "probably high" risk of bias for all key domains AND have most other risk of bias domains as "definitely high" or "probably high."

As indicated in Table 4, using the OHAT Handbook approach resulted in all 40 studies being categorized

as Tier 2. Tier 2 studies have an overall moderate risk of bias and are generally low quality, which decreases

the reliability of their results. In general, most of the studies did not employ an appropriate statistical

approach to assess potential confounding, utilized area-level exposure measurements, and were not able to

assess the temporal relationship between nickel exposure and the outcome of interest, indicating a high risk

of bias in these domains. We note, however, that even though all studies were categorized as Tier 2, there

was variability in the level of risk of bias among studies, with some studies having a higher or lower risk

of bias across more domains than others. The results of studies with a lower risk of bias across the key

domains of exposure assessment, outcome assessment, and confounding are likely more reliable than

studies with a higher risk of bias across these domains.

Below, we evaluate the results of the studies in the context of the factors that contributed to their moderate

risk of bias and generally low quality. We did not fully incorporate study quality considerations into the

discussion of all studies reporting null or negative results, however, because such results are not supportive

of a causal association. Instead, these factors are shown in Tables 1 and 2.


3.3 Studies of Female Reproductive Outcomes

CalOEHHA (2018) reviewed three studies that assessed potential associations between nickel exposure and

female reproductive outcomes. Each study examined different outcomes, so the consistency of findings

across studies cannot be evaluated. Two of the studies reported null associations. Bloom et al. (2011)

reported no association between blood nickel levels and time to pregnancy, and Maduray et al. (2017)

reported no associations between hair or serum nickel concentrations and preeclampsia. Zheng et al. (2015)

evaluated associations between serum nickel concentrations and polycystic ovary syndrome (PCOS), as

well as multiple clinical chemistry parameters, including sex hormone levels. The authors reported

statistically significantly higher serum nickel concentrations in PCOS cases compared to controls, and a

statistically significant decrease in sex hormone binding globulin (SHBG) levels with increasing serum

nickel concentrations. There were no associations between serum nickel concentrations and other clinical

chemistry parameters that would be expected to change in relation to SHBG, however, including estradiol,

testosterone, insulin, glucose, and cholesterol. Given the factors that contributed to the moderate risk of

bias for this study (particularly a lack of accounting for important confounders, as well as likely selection

bias and an inability to assess the temporal relationship between nickel exposure and the outcomes

evaluated), the reported associations between nickel exposure and PCOS or alterations in SHBG levels need

to be confirmed in other studies before they can be considered to support a hazard listing for nickel as a

female reproductive toxicant. Overall, the three studies reviewed by CalOEHHA (2018) do not provide

evidence for a causal association between exposure to nickel compounds and adverse female reproductive

outcomes. As each of these studies measured nickel concentrations in non-occupational participants, none

included exposures to nickel metal.

3.4 Studies of Male Reproductive Outcomes

CalOEHHA (2018) reviewed nine studies evaluating potential associations between nickel exposure and

male reproductive outcomes. None of these studies accounted for important potential confounding

variables, employed appropriate statistical approaches, or were able to assess temporal relationships given

their cross-sectional design. As noted above, one of these studies (Slivkova et al. 2009) did not actually

evaluate associations between nickel exposure and any reproductive outcome, and should not be included

in an evaluation of the potential reproductive toxicity of nickel. The remaining eight studies evaluated

associations between nickel in serum, semen, or urine and the outcomes of circulating hormone levels,

sperm DNA damage, or sperm function parameters. While hormone levels and sperm DNA damage may

or may not be indicative of adverse effects on male reproduction, the sperm function parameters are more

direct indicators of adverse effects, such as infertility.

Zeng et al. (2013) reported no association between urinary nickel levels and plasma testosterone, whereas

Sancini et al. (2014) reported a statistically significant decrease in plasma testosterone with increasing

urinary nickel levels. Wang et al. (2016) reported a statistically significant decrease in the ratio of

testosterone to luteinizing hormone (LH) with increasing urinary nickel levels, but no effect of nickel on

levels of testosterone or other sex hormones. The authors also reported no association between urinary

nickel levels and markers of sperm DNA damage (comet assay tail length, distributed moment, and tail

percent). By contrast, Zhou et al. (2016) reported a statistically significant association between increasing

urinary nickel levels and increased comet assay tail length in sperm, but no association with distributed

moment or tail percent. It is notable that the populations studied by Zeng et al. (2013) and Wang et al.

(2016) consisted of male partners in couples undergoing infertility assessment in China, and those studied

by Zhou et al. (2016) were infertile Chinese men; thus, the results of these studies are not generalizable to

the general US population.


Regarding associations between nickel exposure and sperm functional parameters, two studies reported

statistically significant decreases in sperm motility (Danadevi et al., 2003; Zafar et al., 2015), whereas two

others reported no effects on motility (Skalnaya et al., 2015; Zeng et al., 2015). Danadevi et al. (2003) also

reported significantly decreased sperm vitality associated with nickel exposure, whereas Skalnaya et al.

(2015) reported no effect of nickel on vitality. Results for sperm head abnormalities were also inconsistent

between studies (Danadevi et al., 2003; Zeng et al., 2015). Skalnaya et al. (2015) reported a statistically

significant association between semen nickel levels and decreased ejaculate volume using the Mann-

Whitney U test, but a correlation analysis did not confirm these results. Zafar et al. (2015) reported

significantly decreased sperm count in association with nickel exposure, whereas the studies by Danadevi

et al. (2003), Skalnaya et al. (2015), and Zeng et al. (2015) found no association between nickel and sperm

count. The populations studied by Zafar et al. (2015) and Zeng et al. (2015) consisted of male partners in

couples undergoing infertility assessment in Pakistan and China, respectively. In addition, the studies by

Zafar et al. (2015) and Skalnaya et al. (2015) used inappropriate statistical methods, reducing the reliability

of their results.

Overall, the results for each of the male reproductive outcomes examined were inconsistent across studies,

and do not support a causal association with nickel exposure. More importantly, because all the studies

were found to have a moderate risk of bias, the validity of their results is questionable, and they do not

support a hazard listing for nickel or nickel compounds as male reproductive toxicants.

3.5 Studies of Developmental Outcomes

CalOEHHA (2018) reviewed 28 studies evaluating potential associations between nickel exposure and

various developmental outcomes, including birth defects, low birth weight, adverse pregnancy outcomes,

autism spectrum disorder (ASD), early-life cancers, and DNA damage. Some of these studies had a lower

risk of bias across more key domains than others, so their results may be more reliable than studies with a

higher risk of bias across more key domains, as discussed below.

Birth Defects

Of the eight studies of nickel associations with birth defects reviewed by CalOEHHA (2018), one (Huang

et al., 2011) did not evaluate statistical associations between nickel exposure and any health outcome, and

should not be included in an evaluation of the potential developmental toxicity of nickel. Of the remaining

seven studies, two reported statistically significant associations with birth defects (Chashschin et al., 1994;

Zheng et al., 2012), whereas four reported no associations between nickel exposure and birth defects (Friel

et al., 2005; Vaktskjold et al., 2006, 2008b; Manduca et al., 2014) and one reported a statistically significant

negative (i.e., protective) association with nickel exposure (Yan et al., 2017). It is important to note that

one of the positive studies (Chashschin et al., 1994) includes a disclaimer from the journal editors stating

the following: "Although the results are incompletely documented and thus must be considered

inconclusive, they identify a concern that requires a more comprehensive and quantitative epidemiologic

investigation." In addition, the cohort of female workers studied by Chashschin et al. (1994) was

subsequently investigated more thoroughly by Vaktskjold et al. (2006, 2008b), who did not reproduce the

earlier positive findings.

The studies by Vaktskjold et al. (2006, 2008b) measured concentrations of water-soluble nickel in aerosols

at a Russian nickel refinery, in conjunction with urinary nickel concentrations, to estimate low and high

nickel exposure categories. Similarly, Chashschin et al. (1994) measured water-soluble nickel sulfate

aerosol concentrations in two specific areas of the refinery. Other nickel compounds with potential

exposures in the refinery, such as insoluble forms of nickel, were not discussed and therefore not accounted

for in these exposure assessments. Regardless, the studies by Vaktskjold et al. (2006, 2008b) reported no


associations between nickel exposures that were more than 1,000-fold higher than in ambient air and birth

defects.

Although all the studies of nickel associations with birth defects have a moderate risk of bias because of

inappropriate statistical methods and a lack of accounting for important confounders, three of the studies

reporting null or negative associations assessed exposures and outcomes using well-established methods,

had low probability for both selection and attrition bias, and were designed to directly assess temporality

of exposure and outcome, increasing the reliability of their results (Vaktskjold et al., 2006, 2008b; Yan et

al., 2017). Because the majority of the studies reported null or negative results (including those with more

reliable results), they do not support a causal association between exposure to nickel and/or nickel

compounds and birth defects.

Low Birth Weight

Ten studies evaluated nickel associations with measures of low birth weight. Four of these studies reported

statistically significant, positive associations between nickel exposure and lower birth weight (Bell et al.,

2010; Ebisu and Bell, 2012; Basu et al., 2014; Laurent et al., 2014), and one study reported a borderline

statistically significant association with lower birth weight and a significant association with decreased

newborn head circumference (Pederson et al., 2016). Four studies reported no associations between nickel

exposure and birth weight (Odland et al., 1999, 2004; McDermott et al., 2014; Hu et al., 2015), and one

study reported a negative association between nickel exposure and low birth weight (Vaktskjold et al.,

2007).

Given that the studies by Bell et al. (2010), Ebisu and Bell (2012), and Vaktskjold et al. (2007) have a

lower risk of bias across more domains than the other studies evaluating birth weight, their results are likely

more reliable. However, an independent analysis indicates that the study by Vaktskjold et al. (2007) had

adequate statistical power to detect the effects on low birth weight reported by Ebisu and Bell (2012) (if

they are indeed causal) at nickel exposure concentrations 40-fold lower than those estimated for the workers

in the study (see S. Seilkop comments, submitted separately), indicating the importance of testing

hypotheses generated by univariate analyses in multi-pollutant studies such as those by Ebisu and Bell

(2012) and Bell et al. (2000). Given this analysis, as well as the inconsistency of the results for low birth

weight across studies (even those of similar reliability), the studies evaluating nickel associations with low

birth weight do not provide evidence to support a causal association.

Autism Spectrum Disorder

Three studies evaluated associations between nickel exposure and ASD prevalence, with one reporting no

association (Kalkbrenner et al., 2010) and two studies reporting statistically significant associations

(Windham et al., 2006; Roberts et al., 2013). The latter two studies are limited by a lack of confidence in

the exposure assessment, a lack of accounting for important confounders, and inappropriate statistics, so it

is unclear whether nickel exposures were adequately measured and the positive results are likely attributable

to bias or confounding. Together, these three studies do not provide evidence for a causal association

between nickel exposure and ASD.

DNA Oxidative Damage and Early-Life Cancers

One study reported a statistically significant association between nickel exposure and DNA oxidative

damage in umbilical cord plasma (Ni et al., 2014). Three other studies evaluated risks of early-life cancers,

reporting no associations between nickel exposure and development of neuroblastoma (Heck et al., 2013)

or testicular germ cell tumors (Togawa et al., 2016), and a statistically significant association between

nickel exposure and increased risk of retinoblastoma (Heck et al., 2015). As the outcomes of DNA


oxidative damage and retinoblastoma risk were only evaluated in one study each, and these studies are

limited by several factors including a lack of accounting for confounders, further studies of potential

associations between nickel exposure and these outcomes are needed before they can be considered as

supportive evidence for a causal association.

Adverse Pregnancy Outcomes

Four studies evaluated adverse pregnancy outcomes (including spontaneous abortion and premature birth)

and three reported no associations with nickel exposure (Vaktskjold et al., 2008a; Zheng et al., 2014;

Manduca et al., 2014), whereas one reported an increased risk of spontaneous abortion in nickel-exposed

female workers (Chashschin et al., 1994). As noted above, the study by Chashschin et al. (1994) was not

well documented, and a more recent study of the same cohort did not reproduce the positive findings for

spontaneous abortion (Vaktskjold et al., 2008a). One study reported no association between nickel

exposure and pneumonia in early life (Fuertes et al., 2014). As each of these studies evaluated different

outcomes (with the exception of the inconsistent results for spontaneous abortion in two studies), they do

not provide strong evidence for or against causal associations with nickel exposure.

3.6 Summary

Overall, the results of a risk-of-bias analysis and study evaluation indicate that the epidemiology studies

reviewed by CalOEHHA (2018) do not provide evidence that nickel and/or nickel compounds present a

reproductive or developmental hazard. All reviewed studies had a moderate risk of bias, indicating

generally low quality, due to the lack of appropriate statistical approaches to assess potential confounding,

the use of area-level exposure measurements, and an inability to assess the temporal relationship between

nickel exposure and the outcome of interest. The majority of studies evaluated exposures to soluble and

oxidic forms of nickel (i.e., in ambient air or welding fumes), and the results across the various reproductive

and developmental outcomes examined were largely inconsistent or null. Workers in the refinery studies

had additional exposures to sulfidic and metallic nickel, and the results of these studies were largely null or

not reproducible in more reliable studies. Overall, the epidemiology studies do not provide clear evidence

for associations between any form of nickel and any particular reproductive or developmental outcome.


4 Conclusions

In its review of the evidence for hazard identification for nickel compounds, CalOEHHA (2018) did not

follow a systematic approach and gave more weight to positive studies than those reporting null or negative

associations, regardless of study quality or risk of bias. This led to an evaluation that does not represent the

state of the science regarding the potential reproductive and developmental toxicity of nickel compounds.

This will also limit the ability of DARTIC to form scientifically defensible opinions regarding the

reproductive hazard potential of nickel compounds, making it difficult for DARTIC to determine whether

nickel and nickel compounds meet the CalOEHHA criteria for listing as a known reproductive toxicant.

In a risk-of-bias analysis of the epidemiology studies reviewed by CalOEHHA (2018) using the NTP OHAT

Risk of Bias Rating Tool, we found that all studies have an overall moderate risk of bias, indicating

generally low quality, due to the lack of appropriate statistical approaches to assess potential confounding,

the use of area-level exposure measurements in many studies, and an inability to assess the temporal

relationship between nickel exposure and the outcome of interest. Results across the various reproductive

and developmental outcomes examined were largely inconsistent or null, with no clear evidence for

associations between any form of nickel and any particular outcome. Overall, the epidemiology studies do

not provide evidence that nickel or nickel compounds present a reproductive or developmental hazard.


References

Agency for Toxic Substances and Disease Registry (ATSDR). 2005. "Toxicological Profile for Nickel

(Final)." 397p. August.

Basu, R; Harris, M; Sie, L; Malig, B; Broadwin, R; Green, R. 2014. "Effects of fine particulate matter and

its constituents on low birth weight among full-term infants in California." Environ. Res. 128:42-51.

Bell, ML; Belanger, K; Ebisu, K; Gent, JF; Lee, HJ; Koutrakis, P; Leaderer, BP. 2010. "Prenatal exposure

to fine particulate matter and birth weight: Variations by particulate constituents and sources."

Epidemiology 21(6):884-891. doi: 10.1097/EDE.0b013e3181f2f405.

Bloom, MS; Louis, GM; Sundaram, R; Kostyniak, PJ; Jain, J. 2011. "Associations between blood metals

and fecundity among women residing in New York State." Reprod. Toxicol. 31(2):158-163.

California Office of Environmental Health Hazard Assessment (CalOEHHA). 1993. "Criteria for

Recommending Chemicals for Listing as "Known to the State to Cause Reproductive Toxicity."" 6p.,

November. Accessed on August 22, 2018 at https://oehha.ca.gov/media/downloads/proposition-

65/proposition-65/dartcriterianov1993.pdf.

California Office of Environmental Health Hazard Assessment (CalOEHHA). 2012. "Nickel Reference

Exposure Levels (Final)." 194p., February. Accessed on August 15, 2013 at

http://www.oehha.ca.gov/air/chronic_rels/pdf/032312NiREL_Final.pdf.

California Office of Environmental Health Hazard Assessment (CalOEHHA). 2018. "Evidence on the

Developmental and Reproductive Toxicity of Nickel and Nickel Compounds." Reproductive and Cancer

Hazard Assessment Branch. 347p., July. Accessed on August 1, 2018 at

https://oehha.ca.gov/media/downloads/crnr/nihid072718.pdf.

Chashschin, VP; Artunina, GP; Norseth, T. 1994. "Congenital defects, abortion and other health effects in

nickel refinery workers." Sci. Total Environ. 148(2-3):287-291.

Danadevi, K; Rozati, R; Reddy, PP; Grover, P. 2003. "Semen quality of Indian welders occupationally

exposed to nickel and chromium." Reprod. Toxicol. 17(4):451-456. doi: 10.1016/S0890-6238(03)00040-

6.

De Brouwere, K; Buekers, J; Cornelis, C; Schlekat, CE; Oller, AR. 2012. "Assessment of indirect human

exposure to environmental sources of nickel: Oral exposure and risk characterization for systemic effects."

Sci. Total Environ. 419:25-36.

Ebisu, K; Bell, ML. 2012. "Airborne PM2.5 chemical components and low birth weight in the northeastern

and mid-Atlantic regions of the United States." Environ. Health Perspect. 120(12):1746-1752. doi:

10.1289/ehp.1104763.

European Food Safety Authority (EFSA) Scientific Committee. 2017. "Guidance on the use of the weight

of evidence approach in scientific assessments." EFSA J. 15(8):4971. doi: 10.2903/j.efsa.2017.4971.


Friel, JK; Longerich, H; Jackson, SE; Pushpananthan, C; Wright, JR. 2005. "Possible altered mineral

metabolism in human anencephalic fetuses." Nutr. Res. 25:103-109.

Fuertes, E; MacIntyre, E; Agius, R; Beelen, R; Brunekreef, B; Bucci S; Cesaroni, G; Cirach, M; Cyrys, J;

Forastiere, F; Gehring, U; Gruzieva, O; Hoffmann, B; Jedynska, A; Keuken, M; Klumper, C; Kooter, I;

Korek, M; Kramer, U; Molter, A; Nieuwenhuijsen, M; Pershagen, G; Porta, D; Postma, DS; Simpson, A;

Smit, HA; Sugiri, D; Sunyer, J; Wang, M; Heinrich, J. 2014. "Associations between particulate matter

elements and early-life pneumonia in seven birth cohorts: Results from the ESCAPE and TRANSPHORM

projects." Int. J. Hyg. Environ. Health 217:819-829.

Goodman, JE; Prueitt, RL; Dodge, DG; Thakali, S. 2009. "Carcinogenicity assessment of water-soluble

nickel compounds." Crit. Rev. Toxicol. 39:365-417.

Goodman, JE; Prueitt, RL; Thakali, S; Oller, AR. 2011. "The nickel ion bioavailability model of the

carcinogenic potential of nickel-containing substances in the lung." Crit. Rev. Toxicol. 41(2):142-174.

Heck, JE; Park, AS; Qiu, J; Cockburn, M; Ritz, B. 2013. "An exploratory study of ambient air toxics

exposure in pregnancy and the risk of neuroblastoma in offspring." Environ. Res. 127:1-6.

Heck, JE; Park, AS; Qiu, J; Cockburn, M; Ritz, B. 2015. "Retinoblastoma and ambient exposure to air

toxics in the perinatal period." J. Expo. Sci. Environ. Epidemiol. 25:182-186.

Hu, X; Zheng, T; Cheng, Y; Holford, T; Lin, S; Leaderer, B; Qiu, J; Bassig, BA; Shi, K; Zhang, Y; Niu, J;

Zhu, Y; Li, Y; Guo, H; Chen, Q; Zhang, J; Xu, S; Jin, Y. 2015. "Distributions of heavy metals in maternal

and cord blood and the association with infant birth weight in China." J. Reprod. Med. 60:21-29.

Huang, J; Wu, J; Li, T; Song, X; Zhang, B; Zhang, P; Zheng, X. 2011. "Effect of exposure to trace elements

in the soil on the prevalence of neural tube defects in a high-risk area of China." Biomed. Environ. Sci.

24(2):94-101. doi: 10.3967/0895-3988.2011.02.002.

Kalkbrenner, AE; Daniels, JL; Chen, J-C; Poole, C; Emch, M; Morrissey, J. 2010. "Perinatal exposure to

hazardous air pollutants and autism spectrum disorders at age 8." Epidemiology. 21:631-641.

Laurent, O; Hu, J; Li, L; Cockburn, M; Escobedo, L; Kleeman, MJ; Wu, J. 2014. "Sources and contents of

air pollution affecting term low birth weight in Los Angeles County, California, 2001-2008." Environ. Res.

134:488-495.

Maduray, K; Moodley, J; Soobramoney, C; Moodley, R; Naicker, T. 2017. "Elemental analysis of serum

and hair from pre-eclamptic South African women." J. Trace Elem. Med. Biol. 43:180-186.

Manduca, P; Naim, A; Signoriello, S. 2014. "Specific association of teratogen and toxicant metals in hair

of newborns with congenital birth defects or developmentally premature birth in a cohort of couples with

documented parental exposure to military attacks: Observational study at Al Shifa hospital, Gaza,

Palestine." Int. J. Environ. Res. Public Health 11:5208-5223.

McDermott, S; Bao, W; Aelion, CM; Cai, B; Lawson, AB. 2014. "Does the metal content in soil around a

pregnant woman's home increase the risk of low birth weight for her infant?" Environ. Geochem. Health

36(6):1191-1197. doi: 10.1007/s10653-014-9617-4.


National Toxicology Program (NTP). 2015a. "Handbook for Conducting a Literature-Based Health

Assessment Using OHAT Approach for Systematic Review and Evidence Integration." Office of Health

Assessment and Translation (OHAT). 98p., January 9. Accessed on January 12, 2015 at

http://ntp.niehs.nih.gov/pubhealth/hat/noms/index-2.html.

National Toxicology Program (NTP). 2015b. "OHAT Risk of Bias Rating Tool for Human and Animal

Studies." Office of Health Assessment and Translation (OHAT). 37p., January. Accessed on January 12,

2015 at http://ntp.niehs.nih.gov/pubhealth/hat/noms/index-2.html.

Ni, W; Huang, Y; Wang, X; Zhang, J; Wu, K. 2014. "Associations of neonatal lead, cadmium, chromium

and nickel co-exposure with DNA oxidative damage in an electronic waste recycling town." Sci. Total

Environ. 472:354-362.

Odland, JO; Nieboer, E; Romanova, N; Thomassen, Y; Norseth, T; Lund, E. 1999. "Urinary nickel

concentrations and selected pregnancy outcomes in delivering women and their newborns among arctic

populations of Norway and Russia." J. Environ. Monit. 1(2):153-161.

Odland, JO; Nieboer, E; Romanova, N; Thomassen, Y. 2004. "Urinary nickel concentrations and selected

pregnancy outcomes in delivering women and their newborns among arctic populations of Norway and

Russia." Int. J. Circumpolar Health 63(2):169-187.

Pedersen, M; Gehring, U; Beelen, R; Wang, M; Giorgis-Allemand, L; Andersen, A-MN; Basagana, X;

Bernard, C; Cirach, M; Forastiere, F; de Hoogh, K; Grazulevicviene, R; Gruzieva, O; Hoek, G; Jedynska,

A; Klumper, C; Kooter, IM; Kramer, U; Kukkonen, J; Porta, D; Postma, DS; Raaschou-Nielsen, O; van

Rossem, L; Sunyer, J; Sorensen, M; Tsai, MY; Vrijkotte, TG; Wilhelm, M; Nieuwenhuijsen, MJ;

Pershagen, G; Brunekreef, B; Kogevinas, M; Slama, R. 2016. "Elemental constituents of particulate matter

and newborn's size in eight European cohorts." Environ. Health Perspect. 124:141-150.

Roberts, AL; Lyall, K; Hart, JE; Laden, F; Just, AC; Bobb, JF; Koenen, KC; Ascherio, A; Weisskopf, MG.

2013. "Perinatal air pollutant exposures and autism spectrum disorder in the children of nurses' health study

II participants." Environ. Health Perspect. 121:978-984.

Sancini, A; De Sio, S; Gioffre, PA; Casale, T; Giubilati, R; Pimpinella, B; Scala, B; Suppi, A; Bonomi, S;

Samperi, I; Rosati, MV; Tomei, G; Tomei, F; Caciari, T. 2014. "Correlation between urinary nickel and

testosterone plasma values in workers occupationally exposed to urban stressors." Annali di igiene:

medicina preventiva e di comunita. 26:237-254.

Skalnaya, MG; Serebryansky, EP; Yurasov, VV; Tinkov, AA; Demidov, VA; Skalny, AV. 2015.

"Association between semen quality and level of 20 essential and toxic metals in ejaculate." Trace Elem.

Electrolytes. 32:126-132.

Slivkova, J; Popelkova, M; Massanyi, P; Toporcerova, S; Stawarz, R; Formicki, G; Lukac, N; Putala, A;

Guzik, M. 2009. "Concentration of trace elements in human semen and relation to spermatozoa quality."

J. Environ. Sci. Health A Tox. Hazard Subst. Environ. Eng. 44:370-375.

Texas Commission on Environmental Quality (TCEQ). 2017. "TCEQ Guidelines for Systematic Review

and Evidence Integration." Toxicology Division. 53p., December 20. Accessed on May 9, 2018 at

https://www.tceq.texas.gov/assets/public/implementation/tox/dsd/whitepaper/srguidelines.pdf.


Togawa, K; Le Cornet, C; Feychting, M; Tynes, T; Pukkula, E; Hansen, J; Olsson, A; Oksbjerg Dalton, S;

Nordby, KC; Uuksulainen, S; Wiebert, P; Woldbaek, T; Skakkebaek, NE; Fervers, B; Schuz, J. 2016.

"Parental occupational exposure to heavy metals and welding fumes and risk of testicular germ cell tumors

in offspring: A registry-based case-control study." Cancer Epidemiol. Biomarkers Prev. 25(10):1426-1434.

US EPA. 1986. "Health Assessment Document for Nickel and Nickel Compounds (Final)." Environmental

Criteria and Assessment Office. Office of Health and Environmental Assessment, EPA-600/8-83-012FF,

438p., September.

US EPA. 2018a. "Systematic Review Protocol for the IRIS Chloroform Assessment (Inhalation) [CASRN

67-66-3] (Preliminary Materials Draft)." EPA/635/R-17/486, 76p., January.

US EPA. 2018b. "Application of Systematic Review in TSCA Risk Evaluations (Final)." Office of

Chemical Safety and Pollution Prevention; US EPA, Office of Pollution Prevention and Toxics. EPA

Document # 740-P1-8001. 248p., May. Accessed on June 15, 2018 at

https://www.epa.gov/sites/production/files/2018-06/documents/final_application_of_sr_in_tsca_05-31-

18.pdf.

Vaktskjold, A; Talykova, LV; Chashchin, VP; Nieboer, E; Thomassen, Y; Odland, JO. 2006. "Genital

malformations in newborns of female nickel-refinery workers." Scand. J. Work Environ. Health 32(1):41-

50. doi: 10.5271/sjweh.975.

Vaktskjold, A; Talykova, LV; Chashchin, VP; Odland, JO; Nieboer, E. 2007. "Small-for-gestational-age

newborns of female refinery workers exposed to nickel." Int. J. Occup. Med. Environ. Health 20(4):327-

338. doi: 10.2478/v10001-007-0034-0.

Vaktskjold, A; Talykova, LV; Chashchin, VP; Odland, JO; Nieboer, E. 2008a. "Maternal nickel exposure

and congenital musculoskeletal defects." Am. J. Ind. Med. 51(11):825-833.

Vaktskjold, A; Talykova, LV; Chashchin, VP; Odland, JO; Nieboer, E. 2008b. "Spontaneous abortions

among nickel-exposed female refinery workers." Int. J. Environ. Res. Public Health 18(2):99-115. doi:

10.1080/09603120701498295.

Wang, YX; Sun, Y; Huang, Z; Wang, P; Feng, W; Li, J; Yang, P; Wang, M; Sun, L; Chen, YJ; Liu, C;

Yue, J; Gu, LJ; Zeng, Q; Lu, WQ. 2016. "Associations of urinary metal levels with serum hormones,

spermatozoa apoptosis and sperm DNA damage in a Chinese population." Environ. Int. 94:177-188.

Windham, GC; Zhang, L; Gunier, R; Croen, LA; Grether, JK. 2006. "Autism spectrum disorders in relation

to distribution of hazardous air pollutants in the San Francisco Bay area." Environ. Health Perspect.

114(9):1438-1444. doi: 10.1289/ehp.9120.

Yan, L; Wang, B; Li, Z; Liu, Y; Huo, W; Wang, J; Li, Z; Ren, A. 2017. "Association of essential trace

metals in maternal hair with the risk of neural tube defects in offspring." Birth Defects Res. 109:234-243.

Zafar, A; Eqani, SA; Bostan, N; Cincinelli, A; Tahir, F; Shah, ST; Hussain, A; Alamdar, A; Huang, Q;

Peng, S; Shen, H. 2015. "Toxic metals signature in the human seminal plasma of Pakistani population and

their potential role in male infertility." Environ. Geochem. Health 37:515-527.

Zeng, Q; Zhou, B; Feng, W; Wang, YX; Liu, AL; Yue, J; Li, YF; Lu, WQ. 2013. "Associations of urinary

metal concentrations and circulating testosterone in Chinese men." Reprod. Toxicol. 41:109-114.


Zeng, Q; Feng, W; Zhou, B; Wang, YX; He, XS; Yang, P; You, L; Yue, J; Li, YF; Lu, WQ. 2015. "Urinary

metal concentrations in relation to semen quality: A cross-sectional study in China." Environ. Sci. Technol.

49:5052-5059.

Zheng, G; Zhong, H; Guo, Z; Wu, Z; Zhang, H; Wang, C; Zhou, Y; Zuo, Z. 2014. "Levels of heavy metals

and trace elements in umbilical cord blood and the risk of adverse pregnancy outcomes: A population-

based study." Biol. Trace Elem. Res. 160(3):437-444. doi: 10.1007/s12011-014-0057-x.

Zheng, G; Wang, L; Guo, Z; Sun, L; Wang, L; Wang, C; Zuo, Z; Qiu, H. 2015. "Association of serum

heavy metals and trace element concentrations with reproductive hormone levels and polycystic ovary

syndrome in a Chinese population." Biol. Trace Elem. Res. 167:1-10.

Zheng, X; Pang, L; Wu, J; Pei, L; Tan, L; Yang, C; Song, X. 2012. "Contents of heavy metals in arable

soils and birth defect risks in Shanxi, China: A small-area level geographical study." Popul. Environ.

33:259-268.

Zhou, Y; Fu, XM; He, DL; Zou, XM; Wu, CQ; Guo, WZ; Feng, W. 2016. "Evaluation of urinary metal

concentrations and sperm DNA damage in infertile men from an infertility clinic." Environ. Toxicol.

Pharmacol. 45:68-73.

G:\Projects\218143_NiPERA_Prop65\WorkingFiles\Gradient_Comments_OEHHA_Nickel_Prop65_09102018.docx

Tables

T-1 G:\Projects\218143_NiPERA_Prop65\WorkingFiles\Gradient_Comments_OEHHA_Nickel_Prop65_09102018.docx

Table 1 Characteristics of Epidemiology Studies Evaluating Nickel Exposure and Reproductive and Developmental Effects that Impact Study Quality

Study Outcome Category

Study Design

Study Population

Temporality

Exposure Assessment

Outcome Ascertainment

Potential Confounders Considered

Statistical Analysis

Selection Bias

Sample Size Personal

Measurement Metric Used Form of Nickel

Statistical Approach

Dose-Response Assessed

Dose-Response Relationship

with Ni

Bloom et al., 2011

Female reproductive

Cohort Population-based

80 Yes Yes Whole blood NR. Blood Ni concentration analyzed using ICP-MS.

Home pregnancy test

Yes Cox proportional

hazards models

Yes Yes

Zheng et al., 2015

Female reproductive

Case-control

High risk, Hospital-

based

201 No Yes Serum NR. Serum Ni concentration analyzed using ICP-MS.

Blood sample No Mann-Whitney U

(PCOS)

Yes Yes

Yes Linear regression (Hormone

levels)

Yes Yes

Maduray et al., 2017

Female reproductive

Case-control


based

66 No Yes Pubic hair; serum

NR. Hair and serum Ni concentration analyzed using ICP-OES.

Medical records

No Mann-Whitney U,

t-test

No NA

Danadevi et al., 2003

Male reproductive

Cross-sectional

High risk, Occupation

114 No Yes Whole blood NR. Blood Ni concentration analyzed using ICP-MS.

Ejaculate No Linear regression

Yes Yes, for progressive motility, tail defects, and

vitality

Slivkova et al., 2009

Male reproductive

Cross-sectional


based

47 No Yes Semen NR. Ni quantification method was not clearly reported.

Ejaculate No r No NA

Zeng et al., 2013

Male reproductive

Cross-sectional


based

118 No Yes Urine, creatinine-adjusted

NR. Urinary Ni concentration analyzed using ICP-MS.

Peripheral blood sample

Yes Linear regression

Yes No

Sancini et al., 2014

Male reproductive

Cross-sectional



NR. Urinary Ni determined by complexation with APDC and atomic absorption analysis in graphite furnace.

Blood sample Yes Linear regression

Yes Yes

Skalnaya et al., 2015

Male reproductive

Cross-sectional

High risk, insufficient information

148 No Yes Semen NR. Semen Ni concentration analyzed using ICP-MS.

Ejaculate No Mann-Whitney U, r

No NA



Study Design

Study Population

Temporality

Exposure Assessment




Selection Bias






with Ni

Zafar et al., 2015

Male reproductive

Cross-sectional

High risk, hospital-

based

75 No Yes Semen NR. Semen Ni concentration analyzed using ICP-MS.

Ejaculate No One-way ANOVA,

r

No NA

Zeng et al., 2015

Male reproductive

Cross-sectional


based



Ejaculate Yes Logistic regression,

Linear regression

Yes Yes, for percent and

sperm abnormal head

Wang et al., 2016

Male reproductive

Cross-sectional


based

551 (serum hormone),

460 (spermatozoa

apoptosis), 516 (sperm

DNA damage)

No Yes Urine, creatinine-adjusted


Semen sample, blood sample


Yes Yes, for some endpoints

Zhou et al., 2016

Male reproductive

Cross-sectional


based



Ejaculate Yes Linear regression

Yes Yes, for Comet tail length

Chashschin et al., 1994

Developmental Cross-sectional


698 No Yes Urine, 24-hour Nickel sulfate aerosols.

Medical records

Yes POR No NA

Odland et al., 1999



265 No Yes Urine NR. Urinary Ni concentration analyzed using electrothermal atomic absorption spectrometry.

Medical records


Yes No

Odland et al., 2004



262 No Yes Blood Urine

Placenta

NR. Medical records

No Linear regression

Yes No

Friel et al., 2005



based

55 No Yes Liver, kidney, diaphragmatic muscle, sciatic

nerve, pancreas

NR. Ni concentration analyzed using ICP-MS.

Medical records

No t-test No NA

Vaktskjold et al., 2006

Developmental Cohort High risk, Occupation

23,141 Yes Yes Employment, urine, air

Water-soluble nickel compounds and solvents.

Birth Registry Yes Logistic regression

Yes No



Study Design

Study Population

Temporality

Exposure Assessment




Selection Bias






with Ni

Windham et al., 2006

Developmental Case-control

Population-based

941 No No 1996 US EPA HAPs data

NR. HAPs concentrations estimated from Gaussian air dispersion model that combines emissions inventories from mobile, point and area sources with data on local meteorology, chemical decay rates, secondary formation, and deposition.

Medical records

Yes Logistic regression

Yes Yes

Vaktskjold et al., 2007



Water-soluble nickel subfraction of respirable aerosol fraction.


Yes Yes

Vaktskjold et al., 2008a





Birth Registry, Questionnaire


Yes No

Vaktskjold et al., 2008b





Yes No

Bell et al., 2010

Developmental Cohort Population-based

76,788 Yes No County-wide exposure

estimates via ambient air

monitors

Ni as an oil-combustion-associated element of PM2.5 determined by X-ray fluorescence.

Birth certificate Yes Logistic regression,

Linear regression

Yes Yes



Study Design

Study Population

Temporality

Exposure Assessment




Selection Bias






with Ni

Kalkbrenner et al., 2010


Population-based

3,212 Yes No US EPA HAPs data (NATA-

1996)

Ni compounds as HAPs. HAPs concentrations estimated from Gaussian air dispersion, based on emissions inventory information for point and area sources as well as data on local meteorology and secondary pollutant formation.

Developmental records


Yes No

Huang et al., 2011

Developmental Ecologic Population-based

NR No No Mixed village soil sample

NR. Soil Ni concentration analyzed using ICP-MS.

Physician verification

NR Poisson regression

Yes Yes

Zheng et al., 2012

Developmental Ecologic Population-based

379 No No Village soil sample

NR. Soil Ni concentration analyzed using ICP-MS.

Birth records Yes Poisson regression

Yes Yes

Ebisu and Bell, 2012


1,207,800 Yes No County-wide exposure

estimates via ambient air

monitors

Ni as an element of PM2.5. Average level of exposure calculated during gestation and each trimester.

Birth certificate Yes Logistic regression

Yes Yes

Heck et al., 2013


Population-based

14,677 Yes No Measurements from nearest ambient air

monitors

Ni as an air toxic.

California Cancer Registry


Yes No

Roberts et al., 2013

Developmental Nested case-

control


22,426 Yes No US EPA HAPs data (NATA-1990, 1996, 1999, 2002)

Ni as a HAP. Questionnaire, Telephone

administration of the ADI-R


Yes Yes

Basu et al., 2014


646,296 Yes No Measurements via ambient air

monitors

Ni as an element of PM2.5.

Birth records Yes Logistic regression,

Linear regression

Yes Yes



Study Design

Study Population

Temporality

Exposure Assessment




Selection Bias






with Ni

Fuertes et al., 2014

Developmental Meta-analysis

of cohorts

Population-based

15,980 Yes Yes LUR estimates at residence

Ni as an element of PM2.5 and PM10. Annual average exposure estimated.

Parental reports


Yes Yes

Laurent et al., 2014


960,945 Yes No 4 × 4 km exposure

estimates via CTM

Ni as an element of primary PM. Simulated PM concentrations calculated for PM2.5 and PM0.1.

Birth certificate Yes Generalized additive models

Yes Yes

Manduca et al., 2014



based

69 Yes Yes Hair NR. Hair Ni concentration analyzed using DRC-ICP-MS.

Medical records

No Wilcoxon-Mann-Whitney

test

No NA

McDermott et al., 2014

Developmental Cohort High risk, Minority

9,920 Yes Yes Kriging soil estimates at

residence

NR. Soil Ni concentration analyzed by an independent analytical laboratory.

Birth certificate Yes Generalized additive models

Yes No

Ni et al., 2014 Developmental Cross-sectional

Population-based

201 No Yes Umbilical cord blood

NR. Blood Ni concentration analyzed using GFAAS.

Plasma sample measurements


Yes Yes

Zheng et al., 2014


Population-based

179 No Yes Umbilical cord blood

NR. Blood Ni concentrations analyzed using ICP-MS.

Medical records

No Mann-Whitney U

No NA

Heck et al., 2015


Population-based

30,704 Yes No Measurements via nearest ambient air

monitor

Ni as an air toxic.

California Cancer Registry


Yes Yes

Hu et al., 2015



based

81 No Yes Maternal and cord blood

NR. Blood Ni concentrations analyzed using ICP-MS.

Medical records


Yes No

Pedersen et al., 2016

Developmental Meta-analysis

Population-based

34,923 Yes Yes (except for

Lithuanian and Swedish cohorts)

LUR estimates at residence (except for

Lithuanian and Swedish cohorts)

Ni as an element of PM2.5 and PM10. Annual average exposure estimated.

Birth records/ parental reports

Yes Logistic regression;

Linear regression

Yes No

Togawa et al., 2016


Population-based

34,376 Yes Yes JEM NR. Ni exposure determined using Nordic JEMs.

Nationwide cancer

registries


Yes No



Study Design

Study Population

Temporality

Exposure Assessment




Selection Bias






with Ni

Yan et al., 2016



based

452 Yes Yes Hair NR. Hair Ni concentration analyzed using ICP-MS.

Medical records


Yes Yes

Notes: ADI-R = Autism Diagnostic Interview - Revised; ANOVA = Analysis of Variance; CTM = Chemical Transport Models; DNA = Deoxyribonucleic Acid; DRC-ICP-MS = Dynamic Reaction Cell Inductively Coupled Plasma Mass Spectrometry; GFAAS = Graphic Furnace Atomic Absorption Spectrometry; HAP = Hazardous Air Pollutant; ICP-MS = Inductively Coupled Plasma Mass Spectrometry; ICP-OES = Inductively Coupled Plasma-Optical Emission Spectrometry; JEM = Job-Exposure Matrix; LUR = Land Use regression; NA = Not Applicable; NATA = National Air Toxics Assessment; Ni = Nickel; NR = Not Reported; PCOS= Polycystic Ovary Syndrome; PM = Particulate Matter; POR = Prevalence Odds Ratio; R = Correlation Coefficient; US EPA = United States Environmental Protection Agency.


Table 2 Results of Epidemiology Studies Evaluating Nickel Exposure and Reproductive and Developmental Effects


Study Design

Sample Size

Exposure Metric

Outcome Assessed

Effect Measure

Unit of Measure

Effect Estimate

95% CI P for Risk Estimates

Bloom et al., 2011

Female reproductive

Cohort 80 Whole blood Time to pregnancy

% change Per IQR increment

-8.6 NR 0.79

Zheng et al., 2015

Female reproductive

Case-control

201 Blood serum PCOS Difference in

medians (µg/L)

NA 0.41* NR 0.000

FSH % change Per ng/mL increment

0.736 -2.784, 4.256 0.681

LH 5.333 -2.201, 12.866 0.164

Estradiol -4.204 -9.654, 1.247 0.13

Prolactin 3.215 -2.841, 9.270 0.296

T 3.168 -2.569, 8.904 0.278

Progesterone -2.025 -9.557, 5.508 0.597

TSH -6.821 -15.960, 2.319 0.143

DHEA-S 3.234 -0.452, 6.919 0.085

SHBG -12.602 -24.083, -1.122 0.032

Fasting insulin 2.655 -2.866, 8.177 0.344

Fasting glucose 0.978 -0.437, 2.393 0.175

Cholesterol 0.783 -1.149, 2.716 0.425

Triglycerides 0.368 -5.853, 6.589 0.907

Low-density lipoprotein cholesterol

1.38 -1.461, 4.221 0.339

High-density lipoprotein cholesterol

0.41 -2.150, 2.971 0.752

Maduray et al., 2017

Female reproductive

Case-control

66 Pubic hair Pre-eclampsia Difference in

medians (µg/g)

NA -1.54* NR 0.85

Serum Difference in

medians (mg/L)

NA -0.12* NR 0.16

114 Blood serum Sperm count -0.352 NR 0.067



Study Design

Sample Size

Exposure Metric

Outcome Assessed

Effect Measure

Unit of Measure

Effect Estimate


Danadevi et al., 2003

Male reproductive

Cross-sectional

Rapid linear progressive

motility

Mean change

Per µg/L increment

-0.381 0.045

Slow/non-linear progressive

motility

0.386 0.042

Nonprogressive motility

0.141 0.474

Immotility 0.007 0.971

Normal morphology

-0.032 0.872

Head defects -0.145 0.462

Mid-piece defects

0.067 0.734

Tail defects 0.485 0.036

Vitality -0.420 0.026

Slivkova et al., 2009

Male reproductive

Cross-sectional

47 Semen Knob-twisted flagellum, separated flagellum,

flagellum torso, broken

flagellum, retention of cytoplasmic

drop, acrosomal changes, large heads, small

heads, flagellum ball, and other

pathological forms.

r NA NR NA NR

Zeng et al., 2013

Male reproductive

Cross-sectional

118 Urine, creatinine-adjusted

Plasma testosterone

Mean change (ng/dL)

1st quartile REF 0.14#

2nd quartile -0.86 -81.25, 79.53

3rd quartile -83.79 -163.85, -3.74

4th quartile -36.35 -116.31, 43.61



Study Design

Sample Size

Exposure Metric

Outcome Assessed

Effect Measure

Unit of Measure

Effect Estimate


Sancini et al., 2014

Male reproductive

Cross-sectional


Plasma testosterone

Mean log change (ng/mL)

Per unit increment

(log)

-0.466 NR 0.000

Skalnaya et al., 2015

Male reproductive

Cross-sectional

148 Semen Ejaculate volume 0.05

Total sperm count < 39 ×106

Mann-Whitney

U

NA NR NA 0.452

r -0.069 >0.05

Sperm count < 15 × 106 per 1

mL

Mann-Whitney

U

NA NR NA 0.211

r 0.005 >0.05

Progressive sperm motility <

32%

Mann-Whitney

U

NA NR NA 0.708

r -0.041 >0.05

Sperm vitality < 58

Mann-Whitney

U

NA NR NA 0.872

r -0.049 >0.05

Zafar et al., 2015

Male reproductive

Cross-sectional

75 Semen Sperm count r NA -0.26 NA



Study Design

Sample Size

Exposure Metric

Outcome Assessed

Effect Measure

Unit of Measure

Effect Estimate


3rd quartile 0.77 0.43, 1.39

4th quartile 0.67 0.37, 1.02

Sperm count OR 1st quartile REF 0.55#

2nd quartile 1.1 0.39, 3.12

3rd quartile 0.84 0.28, 2.48

4th quartile 0.79 0.27, 2.30

Sperm normal morphology

% change 1st quartile REF 0.86#

2nd quartile 2.02 0.14, 3.90

3rd quartile 0.89 -0.99, 2.76

4th quartile 0.22 -1.66, 2.10

Percent abnormal head


2nd quartile -1.65 -3.90, 0.60

3rd quartile -1.65 -1.32, 3.16

4th quartile -1.65 -0.57, 3.92

Sperm abnormal Head


2nd quartile -1.62 -3.91, 0.67

3rd quartile 1.13 -1.27, 3.53

4th quartile 2.41 -0.09, 4.91

Wang et al., 2016

Male reproductive

Cross-sectional


Estradiol % change Quartiles NR NR 0.98#

FSH % change Quartiles 0.10#

LH % change Quartiles 0.50#

SHBG % change Quartiles 0.86#

Total T % change Quartiles 0.30#

Total T/LH ratio % change Quartiles 0.02#

Total T/LH ratio Mean change

Per µg/L increment

(log)

0.003

Total T/LH ratio (co-adjusted for multiple metals)


2nd quartile -1.7 -16, 13

3rd quartile -8.3 -25, 6.2

4th quartile -14 -32, 2



Study Design

Sample Size

Exposure Metric

Outcome Assessed

Effect Measure

Unit of Measure

Effect Estimate



Annexin V+/PI- spermatozoa

% change Quartiles NR NR 0.10#

PI+ spermatozoa % change Quartiles 0.98#

Annexin V-/PI- spermatozoa

% change Quartiles 0.18#

Annexin V+/PI- spermatozoa

(co-adjusted for multiple metals)

% change 1st quartile REF 0.002

% change 2nd quartile -6.2 -30, 15

% change 3rd quartile 14 -7.3, 39

% change 4th quartile 28 5.1, 55


Comet tail percent

% change Quartiles NR NR 0.81#

Comet tail length


Comet tail distributed

moment


Zhou et al., 2016

Male reproductive

Cross-sectional


Comet tail percent DNA

Mean change

Quartiles NR NR 0.13#

Comet tail length

Mean change

(μm)

Quartiles 0.02#

Comet tail distributed

moment

Mean change

(μm)

Quartiles 0.78#

Comet tail length

Mean change

(μm)

1st quartile REF 0.049#

2nd quartile -0.58 -2.88, 1.71

3rd quartile -0.36 -2.71, 1.99

4th quartile 2.95 0.34, 5.56

Chashschin et al., 1994

Developmental C

Documents

Comment by Gradient on Comment Submissions - Availability ... · malformations, autism spectrum disorders) for tabulation of the nickel exposure levels and study results, with no